Posts
Comments
Oops, yep.
You currently have to message an admin for that. Mostly because we are too worried about people spamming things if we allow them to do this automatically.
Yeah, that is one of my favorite videos by 3Blue1Brown and more like it would be pretty good.
Fixed the image. The problem was that you used a link to the page where imgur embeds the image, and not an actual link to the image itself. To get the image link, right-click on the image and click "Copy Image Address".
Do you know of any videos on this? Ideally while the person is narrating their thoughts out loud.
Sadly Google Docs really doesn't like being hosted on external websites, so I would try to upload them somewhere else (imgur works well for me).
See this for some commentary on that study (and others)
You are missing a few posts that were posted earlier than February 20th. Here are some that I could find:
February 11th by Jim: https://www.lesswrong.com/posts/h4GFHbhxE2pfiadhi/will-ncov-survivors-suffer-lasting-disability-at-a-high-rate
February 5th by willbradshaw: https://www.lesswrong.com/posts/ztDYsD4v7AaAbWEDM/some-quick-notes-on-hand-hygiene
January 29th on the EA Forum: https://forum.effectivealtruism.org/posts/g2F5BBfhTNESR5PJJ/concerning-the-recent-2019-novel-coronavirus-outbreak
I am pretty confident there were also some more comments that were earlier than Feb 20th, but I can't easily find them right now.
I also would be somewhat hesitant about saying that the markets crashed on February 20th. The market continued crashing for quite a while, and this is when Wei Dai wrote some comments about his investment strategy, which, if you had followed it at that point would have still made you a good amount of money.
However, I think your image for the S&P 500 is the wrong image. Will replace it with the right one.
Fixed your embedded images for you.
Or an error in the editorial process that for some reason people are doubling down on. I do think that's a serious option.
Fore previous discussion see:
- https://www.lesswrong.com/posts/h4vWsBBjASgiQ2pn6/credibility-of-the-cdc-on-sars-cov-2
- https://www.lesswrong.com/posts/7K8fSFTnpfN4EBuZ8/how-useful-are-masks-during-an-epidemic
- https://www.lesswrong.com/posts/tREnA5EvTikWf9Mfw/why-telling-people-they-don-t-need-masks-backfired
- https://www.lesswrong.com/posts/H5rr38RASexMwmWCQ/ssc-face-masks-much-more-than-you-wanted-to-know
- https://www.lesswrong.com/posts/B9qzPZDcPwnX6uEpe/coronavirus-justified-practical-advice-summary
- https://www.lesswrong.com/posts/LwcKYR8bykM6vDHyo/coronavirus-justified-practical-advice-thread
I think this is wrong. I've heard of multiple people who have reached out to the authors and illustrators for the article, who have said that the data is indeed correct, but wasn't published in the survey. Here is the relevant tweet response:
Sarah Constantin's overview: https://srconstantin.github.io/2020/03/31/disinfectants.html
Promoted to curated: I generally want to make sure to not spam all of our curated posts with coronavirus related stuff, but this post is good enough to add to our current curation list. I think it analyzes a really important consideration, that has pretty direct implications for many people reading LessWrong, and does so in a transparent and understandable way.
I expect there are probably quite a few errors in this analysis, just because that's how it goes when people try to write things like this quickly and without a lot of domain expertise, but I am pretty confident the overall point is at the very least important to consider.
Yep, I think these three perspectives roughly cover why I think it might have been a good idea. I also think that a good number of people we now think of as having had a large impact on x-risk and who were kind of similar to rationalists (e.g. some of the Manhattan Project scientists) had that impact because they participated in that effort (and the followup cold-war period) for roughly the three reasons you cite.
Yeah, I agree and I do apologize deeply for things not being as fast as they should be.
Yeah, the whole platform is definitely still pretty buggy.
My experience was great after 30 initial minutes of chaos, and then I had an hour of pretty good conversation.
Respond to this comment if you want to ask a question that will be asked in the Q&A section.
This is the fixed link: https://www.youtube.com/watch?v=4qYNjZZGLDQ&feature=youtu.be
Promoted to curated: I've been thinking about this post a lot since it has come out, and it is just a really good presentation of all the research on AI Alignment via debate. It is quite long, which has made me hesitant about curating it for a while, but I now think curating it is the right choice. I also think while it is reasonably technical, it's approachable enough that the basic idea of it should be understandable by anyone giving it a serious try.
Hey, I've been thinking of curating this post. Would you be up for crossposting it's full content now that it's been out for a while, which would allow us to curate it?
Oops, fixed. It’s noon PDT.
I think the model uses a much shorter time for active infections than 2.5 weeks. Not sure what it is, but I think it's closer to 5 days or something like that, which seems to actually fit the behavior of the disease best, on a broad scale.
Agree that it looks weird. I've asked the authors of the project to add a cumulative graph, which makes these assumptions a lot clearer.
That’s active infections. That number corresponds to something like 70% of the population having been infected at some point.
This problem cannot be cleanly solved analytically (because f is discontinuous and obviously lacking a clean closed form), but is expressed by a beautiful and simple differential equation.
Huh, this is a great example. My historical relationship to differential equations has been mostly from the perspective of "it's a method by which you eventually arrive at good analytic descriptions of systems". I think this example really concretely illustrated why that's a wrong perspective.
I expect to learn some lessons from it, in particular in how to design high-density UIs in a LessWrong context, that I expect will show up in other parts of LessWrong. I can also imagine that at some point we would more generally support complicated tables and databases on LessWrong, but that seems quite a while away.
Promoted to curated: This post is a bit more technical than the usual posts we curate, but I think it is still quite valuable to read for a lot of people, since it's about a topic that has already received some less formal treatment on LessWrong.
I also am very broadly excited about trying move beyond a naive bayesianism paradigm, and felt like this post helped me significantly in understanding what that would look like.
Replaced the rot13 in your comment with spoiler tags. (Accessible via >! in the WYSIWYG editor)
Wow, exciting!
See this comment thread: https://www.lesswrong.com/posts/F3q7eL7pdQqhWFTYh/what-should-we-do-once-infected-with-covid-19?commentId=yei98MRd4cjqcDYEp
Alas. Could you briefly link to the other places that have conveyed the ideas sufficiently well for your tastes?
Some comments from Tara Mac Aulay on masks, in the context of donating to hospitals:
Cross posting my comment from elsewhere re certification standards - KN95 is not sufficient in many settings, but still useful in some. Alibaba has a lot of masks available that are unsuitable, and not many masks that are. Finding the masks that are suitable and checking if the factory has been certified under the right standards is very useful
Alibaba suppliers commonly have KN95 masks in stock, but not surgical N95 masks.
In this time of shortage, many hospitals or healthcare workers will take what they can get, but there are some serious differences.
My understanding as is follows:
KN95 masks are tested for particle filtration to a similar standard as NIOSH rated N95 masks in the US. Surigical rated N95 masks certified by NIOSH have to pass a fit test and a fluid test in addition to the filtration tests standard for N95.
KN95 masks are not rated for use in many patient contact settings, they are predominantly designed for industrial use. KN95 masks provide limited protection to the wearer in this clinical setting, some versions also provide limited protection to patients.
In addition to passing the filtration tests, masks suitable for healthcare workers to use during Covid-19 care have additional requirements:
- masks that have a valve are not appropriate for treating suspected cases. Wearing a mask with a valve is much more comfortable, as the valve allows exhaled air to escape without restriction, meaning less force is required to exhale. However, masks with a valve do not protect patients from exposure from HCWs who are potentially infected. Thus masks with a valve should only be used with patients who are confirmed infected, and for some non-clinical staff who have had limited exposure. A mask with a valve is more comfortable for the general population, you'll be able to wear it for longer, and if you're mostly trying to protect yourself rather than others, this is ok. In some clinical settings, a mask with a valve is also ok. If you are buying masks for personal use to *prevent* yourself from becoming sick, get a mask with a valve. If you want to protect others when you are sick, use a mask without a valve.
- because in many settings you'll be using a mask without a valve to protect patients, the inside of the mask will get very damp. If the mask has not been tested for liquid resistance, then this can be a problem, as the fluid may seep through the mask over time, and then again risk infecting patients. If the masks are not tested for this, the filtration efficiency may also be impacted as the mask gets wet. Surgical N95 masks are tested for 2 way fluid resistance (to ensure no fluid from patients comes through the mask to wearer, and to ensure no fluid from wearer comes through to others)
- fit is very important. Surgical N95 masks are tested for fit and escaping/incoming airflow under normal conditions of use. Without this testing, and without a mask that fits well, HCWs and patients can be exposed.
Long story short, KN95 masks are still useful in many settings, and we need those too. But what we need even more is surgical rated N95 masks, ideally certified by NIOSH as such.
It's also likely that many of the masks that are rated as KN95 would pass liquid and fit tests if subjected, but without the testing and certification, I understand why many HCWs would demand certified masks to protect themselves and their families.
She also mentioned that she expects standards in hospitals to decline over time, and that they will eventually accept other mask types for donations, if they don't already.
Note: It's BERI, not BEARI
Gargling (even with water) might be a relatively good intervention: https://pubmed.ncbi.nlm.nih.gov/16242593/
Followup Post: https://srconstantin.github.io/2020/03/20/non-invasive-ventilation.html
It's still available on the Internet Archive: https://web.archive.org/web/20200319023745/https://docs.google.com/document/d/e/2PACX-1vTi-g18ftNZUMRAj2SwRPodtscFio7bJ7GdNgbJAGbdfF67WuRJB3ZsidgpidB2eocFHAVjIL-7deJ7/pub
Huh, that's quite weird. I wonder whether we can ping any of the people at Google we know about this, since I thought that document was quite good, and I linked to it a few times.
Promoted to curated: It took me a really long time to wrap my head around the simulacra-level idea, but after a few weeks of engaging with it passively, it clicked, and it's now something that I've been using quite a bit in my modeling, and have brought up in conversations quite a few times.
This post is still not amazing at explaining the different levels, but it seems better than the posts that came before it. And I do think the concept is important, and worth pointing more people at, so I am curating it.
For all posts to LessWrong, you can subscribe here: https://www.lesswrong.com/feed.xml
If you just want to subscribe to the daily coronavirus link posts, subscribing to just posts by me should do most of the job, which you can do at this link: https://www.lesswrong.com/feed.xml?userId=XtphY3uYHwruKqDyG
I feel torn about whether that will be the most important thing to focus on when it does. On the one hand, when you shut up and multiply, I'm pretty sure that xrisk is many, many times more important. On the other, I really care about people in this community. I've always felt torn about this question of how much extra moral weight to give to those who I care about.
The value of people working on x-risk, from an x-risk perspective, is quite high. So while I sympathize with the conflict in broader terms, in this case, it just seems pretty obvious to me that I care quite a lot about protecting the people in this community, from both a personal and an altruistic perspective.
Traffic is up 30% from last month (which is very significant given that usually most of LessWrong's traffic is driven by the large number of distributed links strewn all around the internet, and so is very stable and already quite high).
We also are being linked from a lot of places. A lot of Facebook links and a lot of Twitter links. I am not sure how to translate Google Sheets visitor numbers into traffic, but we've had pretty consistently 70+ people concurrently on the LessWrong Coronavirus Link Database, which is more than I've seen on basically any other Google Doc linked, and LessWrong, which usually has around 120k unique users a month usually gets around 40 concurrent users. So the additional traffic here seems pretty significant.
I have also heard of at least one higher up place in the UK government which said that they were actively following LessWrong to keep track of the Coronavirus stuff, though this is now a second-degree rumor, so treat this with some salt.
My sense is also that there is a broader ecosystem of people around LessWrong and the rationality community (which includes the EA and SSC communities), which ended up putting a lot of their coronavirus related thoughts here, and who are referring a lot to our content.
I don't think any of this indicates overwhelming amounts of attention on us (which I am mostly glad for), but I do think it indicates attention from a significant subset of smart and informed people that I care a lot about.
Great, looking forward to the post!
My current sense is yes, though I really don't think it's obvious and think that this is a pretty high-stakes call.
My sense is that work in a crisis generally has really high-leverage, and I think there are reasonable arguments that this is the biggest global crisis since World War 2, at least in terms of how the world will be shaped by it, and how much is at stake (you don't see a 30% drop in the stock market that often, and the number of people who will die is quite plausibly more than WW1). I do indeed think that during World War 2 it would have been reasonable for many people on LessWrong to participate in the war effort, and think the same is true in this case. I do think that on an all-things-considered view this is likely going to be much less big of a deal than World War 2, but I think the basic argument is plausible enough that it seems worth betting quite a bit on.
I also think this topic is a much better fit than usual political and news-related topics for LessWrong, because we are ultimately dealing with a "Player vs. Environment" type of threat, and not a "Player vs. Player" type of threat. I think in cases like this, our tools for epistemic rationality and general scientific inquiry are in a good place to shine, and there is less risk of us getting sucked into an adversarial epistemic environment, because the questions to be asked and answered are primarily about pretty stable ground-truths.
The second argument is tractability. I think it's pretty key that people on LessWrong noticed that this was important much earlier than the vast majority of the world, and even the vast majority of the world's intellectual elite. I think this made LessWrong a natural Schelling point of attention, and I don't think it's obvious that a separate Schelling point would emerge if we were to deemphasize coronavirus related topics on LessWrong. This makes me think that LessWrong has at least some responsibility in not damaging the communication around this topic by suddenly deemphasizing the topic, at least not without creating a separate hub where discussion can coalesce instead.
The third argument is that taking a global perspective, I think there is a good argument that you should help out in crises like this, even if working on them is not directly related to your goals, because many other players in the world care a lot about it, and will be deeply grateful for your assistance. I think from a perspective of cooperating with other powers in the world, it's good form to help out with this as much as possible, given the overwhelming importance other people put on this.
The fourth argument is just relevance to all of our wellbeing. I don't think we are at the stage where we can just rely on local governments or standard expert hierarchies to give us advice and tell us what to do. Most governments and municipalities have not yet announced safety measures that seem sufficient to me, and so it's still up to the individual and small communities like LessWrong to figure out what the appropriate level of safety is, and I sadly expect this to be the case for a while.
That mortality estimate seems a bit too low to me, Sarah Constantin estimated 0.1%-0.2% for someone below 40, so the lower end of that seems right to me.
The 0.1% mortality from lack of ventilators also seems too low to me. We might be a bit in luck in Bay Area since I think we are less likely to get it, but you should still expect that if you do get the virus, you get it when everyone else gets it, which means likely overwhelmed hospitals, and 0.1% mortality seems too low for that case. Sarah Constantin also estimates a total fatality rate of 1%-2% when hospitals are overwhelmed.
I don't trust the numbers in the ICL report particularly much, since they are reliably lower than from other reputable sources, and I don't actually see how they got numbers that were that low. I am currently reading through the study that they are basing their estimates on.
All of LessWrong was blocked for about an hour from Facebook, but the issue appears to now be fixed.
Context: https://twitter.com/guyro/status/1240063821974138881
We don't yet have subscription functionality for tags, sadly. I am probably not going to create many non-coronavirus posts for the next few days/weeks, so subscribing just to posts by me should get them basically just these summaries. (To do that, click on the triple-dot at the top of the post when you are logged and select "Subscribe to posts by habryka")
Note: I think it's quite likely this meetup should be cancelled, given the current COVID situation.
Great! Thanks for the update! Will update the spreadsheet.
This is a type of post that should have been vetted with someone for infohazards and harms before being posted, and pending that, I think it should be deleted by moderators or removed by the authors.
As a response to this, the moderator team did indeed reach out (CC'ing David) to one of the people I think David and I both consider to be among the best informed decision-makers in biorisk. With their permission, here is the key excerpt from their response:
> [Me summarizing David:] David is under the impression that people like Elizabeth and Jim are under an obligation to show posts like this to people in biorisk like yourself and definitely not publish if you had any objections (and that posts that don't do so should be immediately deleted). Do you think they are under that obligation and that we should delete posts of this type?
I do not think they are under an obligation to do this. If the post contained object-level nonobvious content related to generating or exacerbating biorisks, I would consider them under a moral obligation to do so, the strength of which would depend on the particulars of the situation.
If the post overemphasizes the degree to which it's handled the outbreak badly only mildly-moderately, or based on reasonable-seeming lines of argumentation in my view, I'd likely consider that within the reasonable range of opinions/perspectives to hold and share on forums like LW. If the post was highly misleading, such that I thought it communicated the wrong picture of the CDC, then I'd think it was epistemically virtuous to make top-level updates, and if the authors refused to do that, writing a counter-post explaining why their post was misleading would seem like a good thing to do to me, though not something I'd want to demand, if I were in position to demand such a thing, which I don't consider myself to be.
Overall, my sense is that you made a prediction that people in biorisk would consider this post an infohazard that had to be prevented from spreading (you also reported this post to the admins, saying that we should "talk to someone who works in biorisk at at FHI, Openphil, etc. to confirm that this is a really bad idea").
We have now done so, and in this case others did not share your assessment (and I expect most other experts would give broadly the same response). I think the authors were correct in predicting a response like this if they had ran it by anyone else, and I also don't think they were under any obligation to run the post by anyone else. This is not in any way a post that is particularly likely to contain infohazards, and I feel very comfortable with people posting posts in this general reference class without running them by anyone else first.
Of course, please continue to point out any errors and ask for factual corrections to the post. And downvote the post if you think it is overall more misleading than helpful. A really big reason for posting things like this publicly is so that we can correct any errors and collectively promote the most important information to our attention. But it seems clear to me that this post does not constitute any significant infohazard that the LessWrong team should prevent from spreading.
I do also think that it is important for LessWrong to have a good infohazard policy, in particular for more object-level ideas, both in biorisk and artificial intelligence. In those domains, I would have probably followed your recommended policy of drafting the post until we had run the post by some more people. I am also happy to chat more with you about what our policies in these more object-level domains should be.
It does seem to me that your comments on this post (and your private messages, and postings to other online groups warning of infohazards in this space) have overall been quite damaging to good discourse norms, and I would strongly request that you stop asking people to take posts down, in particular in the way you have here. Our ability to analyze ideas on the basis of their truth-value, and not the basis of their political competitiveness and implications is one of our core strengths on LessWrong, and it appears to me that in this thread you've at least once argued for conclusions you think are prosocial, but not actually true, which I think is highly damaging.
You've also claimed that hard to access expert-consensus was on your side, when it evidently is not, which I think is also really damaging, since I do think our ability to coordinate around actually dangerous infohazards requires accurate information about the beliefs of our experts, and it seems to me that overall people will walk away with a worse model of that expert consensus after reading your comments.
Most of the consensus that has been built around infohazards in the bio-x-risk community is about the handling of potentially dangerous technological inventions, and major security vulnerabilities. You claimed here (and other places) that this consensus also applied to criticizing government institutions during times of crisis, which I think is wrong, and also has very little chance of actually ever reaching consensus (at least in crises of this type).
The effects of your comments have also been quite significant. The authors of this post have expressed large amounts of stress to me and others. I (and others on the mod team like Ben) have spent multiple hours dealing with this, and overall I expect authority-based criticism like this to have very large negative chilling effects that I think will make our overall ability to deal with this crisis (and others like it) quite a bit worse. You have also continued writing comments like this in private messages and other forums adjacent to LessWrong, with similar negative effects. While I don't have jurisdiction over those places, I can only implore you strongly to cease writing comments of this type, and if you think something is spreading misinformation, to instead just criticize it on the object-level. Here, on LessWrong, where I do have jurisdiction, I still don't think I am likely to invoke my moderator powers, but I am going to strong-downvote any future comments like this (and have already done so for this one).
If you do believe that we should change our infohazard policies to include cases like this, then you are welcome to argue for that by making a new top-level post. But please don't claim that we already have norms, policies and broad buy-in, and that a post like this should have already been taken down, which is just evidently wrong.